Abstract
Glioblastoma (GBM) is one of the most aggressive brain tumors and has a high recurrence rate, and effective treatment is urgently needed. GBM stem cells (GSCs) contribute to GBM recurrence as well as therapeutic resistance to radiation and chemotherapy. Several oncolytic viruses (OVs) have been developed and validated in clinical trials with favorable safety profiles and efficacy against GBM. Recently, Zika virus (ZIKV), a mosquito-borne flavivirus, was shown to preferentially target and kill GSCs and showed promising therapeutic effects in treating GBM in preclinical models. In this review, we summarize the known OVs for the treatment of GBM and highlight the major advantages and existing challenges for the clinical development of oncolytic ZIKV.
Introduction
Brain and other central nervous system (CNS) tumors have a very high fatality rate and are the leading cause of cancer deaths among men younger than 40 years and women younger than 20 years (Siegel et al, 2022). Glioblastoma (GBM) accounts for approximately half of malignant brain tumors and has a low 5-year survival rate of 7% (Miller et al, 2021). The standard treatment for patients with GBM is complete resection followed by radiation and chemotherapy, such as temozolomide (TMZ), whereas the median survival is 1–2 years (Davis, 2016; Miller et al, 2021).
Tumor-treating fields (TTFields) is the fourth treatment option approved by the Food and Drug Administration (FDA) for GBM, along with surgery, radiotherapy, and chemotherapy (Mun et al, 2018). TTFields is a noninvasive antimitotic treatment modality utilizing low-intensity, intermediate-frequency (200 kHz) alternating electric fields to selectively disrupt rapidly dividing GBM cells (Guo et al, 2022; Mun et al, 2018). The effectiveness and safety of TTFields in treating GBM have been demonstrated in a variety of clinical studies in recent years (Guo et al, 2022; Stupp et al, 2017; Stupp et al, 2015). A phase III trial showed that the combination of TTFields and maintenance TMZ chemotherapy resulted in a significant improvement in survival, as the median overall survival was 20.9 months (Stupp et al, 2017). However, almost all GBM patients experience tumor recurrence after standard therapy (Stupp et al, 2017). New treatment is urgently needed to improve survival quality and reduce the recurrence rate.
Several factors contribute to the low cure rate and high recurrence rate of GBM. GBM has complex intra- and intertumoral heterogeneity. Furthermore, recent studies found that 90% of druggable targets changed in recurrent tumor tissues (Louis et al, 2021; Patel et al, 2014; Schafer et al, 2019). The special structure of the blood–brain barrier (BBB) limits drug entry and complicates achievement of therapeutic concentrations (Papademetriou and Porter, 2015; Steeg, 2021). The brain has a special immune system, and the infiltrated lymphocytes of brain tumors are rare and highly immunosuppressed (Ott et al, 2021; Sampson et al, 2020). GBM stem cells (GSCs) are extremely capable of DNA repair to resist radiation and chemotherapy and have the capacity to self-renew and differentiate into stromal and vascular structures, which is responsible for GBM recurrence (Alvarado et al, 2017; Bao et al, 2006; Cheng et al, 2013).
Current therapeutic strategies for GBM show limited improvement in patient survival, reflecting the urgent need to develop more effective treatments. Immunotherapy is a promising new therapeutic strategy, and many types of immunotherapy modalities, including immune checkpoint inhibitors, chimeric antigen receptor T cell therapy, therapeutic vaccination, and oncolytic viruses (OVs), are under investigation in preclinical and clinical trials (Sampson et al, 2020).
OVs include native or genetically modified viruses that selectively replicate in tumor cells and lyse them (Li et al, 2022; Melcher et al, 2021). To date, multiple OVs have reached the clinical trials stage, including herpes simplex virus type 1 (HSV-1), adenovirus, vaccinia, parvovirus, reovirus, poliovirus, and Newcastle disease virus (Kaufman et al, 2015; Shoaf and Desjardins, 2022). Recently, we and others have demonstrated that Zika virus (ZIKV) is a novel oncolytic virus with strong potential for GBM treatment (Chen et al, 2018; Kaid et al, 2018; Zhu et al, 2017). In this brief review, we summarize recent advances in ZIKV as an OV for GBM and highlight existing challenges and perspectives for the clinical development of oncolytic ZIKV.
OVs for GBM and Other Malignant CNS Tumors
Primary gliomas rarely metastasize out of the CNS, therefore they are suitable for local therapies, including OVs. OVs represent a potential immunotherapy for malignant glioma because of their ability to preferentially target and kill tumor cells, and OVs are a good platform to express transgenes, which may improve their tumor cytotoxicity and/or selectivity and modulate the tumor microenvironment (Harrington et al, 2019; Kaufman et al, 2015). Tumor-associated antigens released from lysed tumor cells infected with OVs can activate natural and adaptive antitumor immunity, and danger signals from OVs can also promote the antitumor immune response (Harrington et al, 2019). Several clinical trials focusing on various OVs for the treatment of malignant CNS tumors have been registered at
Potential Oncolytic Viruses for the Treatment of Malignant Central Nervous System Tumors
BBB, blood–brain barrier; CAR, coxsackie adenovirus receptor; dsDNA, double-stranded DNA; dsRNA, double-stranded RNA; HVEM, herpesvirus entry mediator; ssDNA, single-stranded DNA; ss (+) RNA, positive single-stranded RNA; ss (–) RNA, negative single-stranded RNA.
Apart from being cytotoxic to tumor cells, OVs can be engineered to express transgenes to significantly increase their functionality or be combined with other treatment strategies, such as immune checkpoint inhibitors (LaRocca and Warner, 2018). HSV-1-derived medicaments are the most widely tested viral vectors, with several modified strains undergoing clinical trials for patients with high-grade glioma (Conry et al, 2018; Koch et al, 2020; Shoaf and Desjardins, 2022). Talimogene laherparepvec (T-VEC) was the first FDA-approved oncolytic virus for advanced melanoma (Conry et al, 2018). Oncolytic HSV-1 (oHSV-1) armed with interleukin-12 (IL-12) and/or antiprogrammed cell death protein 1 (PD-1) antibody (Omar et al, 2021; Xie et al, 2022; Xu et al, 2017) or combined with radiation or checkpoint inhibitors effectively restrained tumor growth in preclinical experiments, and many ongoing trials are utilizing oHSV-1 for patients with malignant glioma (Bernstock et al, 2021; Friedman et al, 2021; Koch et al, 2020).
Adenovirus is easily attenuated and suitable for encoding large foreign transgenes; thus, it is an attractive vector for clinical development (Kaufman et al, 2015). Multiple oncolytic adenoviruses are undergoing clinical trials in glioma patients with different generation strategies to ensure safety and targeting. Some groups combine other immune activation drugs, including interferon gamma or pembrolizumab (NCT02197169, NCT02798406). Engineered oncolytic adenovirus delivered by neural stem cells (NSCs) was investigated in phase I trials (NCT05139056, NCT03072134).
Poliovirus is a nonenveloped, single-stranded RNA picornavirus that enters cells by binding to CD155. Recombinant nonpathogenic polio–rhinovirus chimera (PVSRIPO) has been generated by replacing the original internal ribosome entry site (IRES) with the IRES from human rhinovirus 2 (Gromeier et al, 2000). PVSRIPO prefers to infect glioma cells, probably due to the upregulation of CD155 on these malignant cells (Mendelsohn et al, 1989; Merrill et al, 2004). Many clinical trials have used PVSRIPO to treat malignant glioma and have shown good safety and efficacy (Desjardins et al, 2018; Suryawanshi and Schulze, 2021).
Oncolytic vaccinia viruses TG6002 expressing the suicide gene FCU1 in combination with the FCU1/5-fluocytosine system has better antitumor effects in multiple human xenograft tumor models, and currently registered trials include phase 1 clinical trials (Foloppe et al, 2019). Oncolytic parvovirus H-1 (ParvOryx), oncolytic Reovirus (REOLYSIN®), and oncolytic Newcastle disease virus (NDV-HUJ, MTH-68/H) have good tolerability and efficacy in subjects suffering from malignant gliomas and have completed phase I/II trials (Csatary et al, 2004; Freeman et al, 2006; Geletneky et al, 2012; Kicielinski et al, 2014).
Despite promising achievements, some special challenges remain in the development of OV treatment for GBM: BBB blockage of drugs and immune cells, the balance between antiviral and antitumor immunity, biosafety considerations and the special immune microenvironment of the CNS, delivery strategies, and tumor heterogeneity, among others (Kaufman et al, 2015; LaRocca and Warner, 2018; Suryawanshi and Schulze, 2021). Nevertheless, OVs have a favorable risk–benefit ratio for patients suffering from GBM and require further refinement.
Oncolytic ZIKV Provides a New Option for GBM Treatment
ZIKV is a mosquito-borne flavivirus belonging to the flavivirus genus, Flaviviridae family, which includes multiple important human pathogens such as dengue virus (DENV), West Nile virus (WNV), Japanese encephalitis virus (JEV), and hepatitis C virus (HCV). ZIKV has an 11-kb positive-stranded RNA genome with positive polarity that encodes three structural proteins and seven nonstructural proteins (Kostyuchenko et al, 2016; Pierson and Kielian, 2013; Sirohi et al, 2016). ZIKV preferentially infects NSCs and neural precursor cells (NPCs) and leads to cell cycle arrest, apoptosis, and differentiation inhibition, which cause serious consequences for brain development (Li et al, 2016; Tang et al, 2016). ZIKV can be transmitted to developing fetuses by crossing the placenta and causes severe CNS anomalies named congenital Zika syndrome (Crooks et al, 2021; Musso et al, 2019; Paixao et al, 2022). In contrast, ZIKV infection in adults is usually symptomless or cause only mild symptoms, such as mild fever and rash, which resolve spontaneously within ∼7 days in less than 20% of patients (Hamel et al, 2015; Petersen et al, 2016).
GSCs and NSCs share several properties and phenotypes and show high expression levels of stem cell markers, including SOX2, OLIG2, and Musashi-1 (MSI1) (Lee et al, 2018). We and others have demonstrated that ZIKV has a particular preference for GSCs and excellent oncolytic effects against GSCs in vitro and in vivo (Chen et al, 2018; Kaid et al, 2018; Zhu et al, 2017). Zhu et al (2017) first provided convincing evidence that ZIKV preferentially infects and kills GSCs with few effects on differentiated glioma cells (DGCs) and normal neural cells. Furthermore, a mouse-adapted strain of ZIKV was confirmed to be able to significantly prolong the survival time of mice with glioma (Zhu et al, 2017). Chen et al (2018) reported that a live-attenuated ZIKV vaccine candidate could prolong the survival time of animals bearing human GSCs and caused no neurological symptoms or behavioral abnormalities.
Kaid et al (2018) verified that the Brazilian ZIKV (ZIKVBR) strain selectively infected human CNS tumor cells rather than human breast, prostate, and colorectal tumor cell lines. Furthermore, Kaid et al (2020) also confirmed the safety and effectiveness of ZIKVBR intrathecal injection in dogs with spontaneous intracranial tumors, which showed the safety of intrathecal ZIKVBR injections in immunocompetent dogs for the first time. Li et al (2019) revealed that ZIKV nonstructural protein 5 (NS5) of the PRVABC59 strain inhibited the proliferation, migration, and invasion of U87 cells and suppressed tumor growth in vivo.
GSCs are the main culprits of drug resistance and recurrence of malignant glioma (Jackson et al, 2019). The tropism of ZIKV for GSCs renders it a very promising oncolytic vector that can kill GSCs and prevent tumor recurrence, while other OVs for malignant glioma showed no preference for GSCs. Other flaviviruses, such as WNV, have no selective effects on GSCs and infect both GSCs and DGCs at high levels (Zhu et al, 2017).
The key mechanism mediating the entry of ZIKV into cells and cell death in human NPCs or GSCs remains controversial. Several studies have shown that anexelekto (AXL) is a potential receptor for ZIKV entry and that blocking AXL reduces ZIKV-induced infection and cell death in NPCs (Nowakowski et al, 2016; Retallack et al, 2016). Zwernik et al (2021) demonstrated that the AXL receptor mediated ZIKV entry into GBM cells in vitro. However, Wells et al (2016) found that genetic deletion of AXL cannot protect human NPCs from ZIKV infection. Srivastava et al (2020) identified neural cell adhesion molecule (NCAM1) as a potential ZIKV receptor by using a time-resolved chemical proteomic strategy to track the early-stage entry of ZIKV into host cells.
Zhu et al (2020) demonstrated that ZIKV's preference for GSCs depends on the SOX2-Integrin avb5 axis. Chavali et al (2017) found that the neural RNA-binding protein MSI1 interacts with the Zika genome and enables viral replication. Both SOX2 and MSI1 are important translational regulators in stem cells and are correlated with the proliferative activity and grade of malignancy in glioma (Dahlrot et al, 2013; Suva et al, 2014). The high levels of SOX2 and MSI1 in GSCs may be the key for their preferential targeting.
Immunological Memory and Combination Therapy with Oncolytic ZIKV
The immune microenvironment of GBM is “cold” due to multiple immunosuppressive mechanisms, and patients have a weak response to immunology (Da Ros et al, 2018; Jackson et al, 2019; Mirzaei et al, 2017; Omuro et al, 2018; Weller et al, 2017). OVs can effectively activate antitumor immunity and increase reactivity to immune checkpoint therapy. ZIKV can reach the CNS by crossing the BBB in adults (da Silva et al, 2017), and Cle et al (2020) demonstrated that ZIKV can cross the BBB in vitro in human BBB models and animal models, which favored leukocyte recruitment. Furthermore, multiple groups reported that ZIKV treatment significantly increased T cell intratumoral infiltration and activation and sensitized GBM to immune checkpoint blockade. Intrathecal ZIKVBR injections in immunocompetent dogs with spontaneous intracranial tumors extended their survival and induced lymphocyte infiltration (Kaid et al, 2020). Crane et al (2020) found that ZIKV could enhance the long-term survival of mice bearing GL261 cells and generate memory T cells.
Multiple combination strategies and arming strategies have shown good safety and efficacy in clinical trials of OVs (Harrington et al, 2019). Oncolytic ZIKV combined with checkpoint inhibitor antibodies demonstrated good therapeutic effects in an animal model (Table 2). Nair et al (2021) demonstrated that the oncolytic activity of ZIKV requires CD8+ T cells and is boosted by anti-PD-1 antibody therapy. Chen et al (2022) revealed that ZIKV promoted activation of the type I interferon signaling pathway of GBM cells and overcame the resistance of GBM to programmed death ligand 1 (PD-L1) blockade in immunocompetent mice with glioma. Sen et al (2022) demonstrated that GSCs with low Schlafen family member 11 (SLFN11) expression are susceptible to ZIKV but resistant to TMZ, serving as a reminder that the combination of OVs and chemotherapy may be a powerful treatment approach. Zhu et al (2017) also confirmed that ZIKV-E218A combined with TMZ had greater antitumor efficacy in vitro. Further preclinical and clinical trials are needed to validate combined strategies and the mechanisms of oncolytic ZIKV.
Important Information on Zika Virus as an Oncolytic Virus
AXL, anexelekto; GSCs, glioblastoma stem cells; NCAM1, neural cell adhesion molecule; NPCs, neural precursor cells; NS5, nonstructural protein 5; NS4B, nonstructural protein 4B; PD-1, programmed cell death 1; PD-L1, programmed cell death ligand 1; SLFN11, Schlafen family member 11; UTR, untranslated region.
Perspective
OVs are promising immunotherapy for malignant glioma. Four OVs have been approved worldwide, including ECHO-7, adenovirus, and two oHSVs. H101, a genetically modified adenovirus, is the world's first oncolytic virus used for head and neck cancer treatment and was approved by Chinese regulators in 2005 (Garber, 2006). T-VEC was the first oHSV approved by FDA for the treatment of metastatic melanoma in 2015 (Otani et al, 2022). Furthermore, the Japan Ministry of Health granted time-limited approval for oHSV-G47Δ for the treatment of malignant glioma in 2021 (Otani et al, 2022). To date, 28 clinical trials of OVs for patients with glioma have been registered and 8 of them have been completed (
Recent studies have confirmed the promising oncolytic effect of ZIKV on GBM and provided significant insight into understanding how ZIKV preferentially infects and kills GSCs. The ability of ZIKV to preferentially infect and kill GSCs provides new hope for a GBM cure and recurrence prevention. However, safety and stability are the main concerns for the clinical application of ZIKV. The live-attenuated vaccine candidate containing a 10-nucleotide deletion in the 3′ untranslated region (Δ10 3′-UTR) of the ZIKV-FSS13025 strain was demonstrated to have good oncolytic effects and safety in BALB/c nude mice bearing human GSCs (Chen et al, 2018). Normal brain cells are almost unaffected in immunocompetent mice treated with the Δ10 3′-UTR of the ZIKV-Dakar strain (Nair et al, 2021). Although ZIKV has demonstrated a good safety profile in the treatment of mice with glioma, safety requires further consideration in clinical applications. Arming with the suicide gene FCU1 to control ZIKV replication may be a viable strategy (Foloppe et al, 2019). A special design is needed to avoid harm to NSCs since both NSCs and GSCs strongly express stem cell markers.
Meanwhile, stably expressing heterologous genes of interest by modifying ZIKV is difficult because it has a relatively small genome size, and the genome organization is monocistronic (Baker and Shi, 2020). Whether radiation and chemotherapy can change the effectiveness and stability of oncolytic ZIKV is unclear. Further mechanistic studies and clinical trials are needed to confirm the efficacy and safety of oncolytic ZIKV and combination strategies for patients suffering from GBM.
Footnotes
Disclosure Statement
No competing financial interests exist.
Funding Information
This work was supported by the National Natural Science Foundation of China (Grant No. 32000129), the National Science Fund for Distinguished Young Scholars (81925025), the Innovative Research Group (81621005) from the NSFC, and the Innovation Fund for Medical Sciences (2019-I2M-5-049) from the Chinese Academy of Medical Sciences.
